Fast.ai Machine Learning lessons on Kaggle kernels

@jcatanza and i have made ML lesson available at Kaggle kernels.

Just fork the notebook and have fun! :slight_smile:

Lesson 1 notebook
Lesson 2 notebook
Lesson 3 notebook
Lesson 4 notebook
Lesson 5 notebook

Let us know if you have any feedback or comments.

5 Likes

just added Lesson 2
there are some errors in the second part of the notebook with the plot_pdp function, i tried this solution but need to do more troubleshooting. github issue log.

However it’s good for video lesson 3.

Thanks @miwojc

@ritika26 for some reason the solution you have there doesn’t work for me. where to you run notebooks? i assume it works there? thanks!

I have created the separate environment for fastai in Anaconda. I am running these notebooks in Jupyter .
It is working for me.

Thank you for quick confirmation @ritika26. it must be due to kernels then. i’ll keep working on it then…

added lesson 3 notebook
also lesson 2 notebook was updated as per latest chages on github repo and now all works :slight_smile:

thanks to @jcatanza for adding lessons 4 and 5! now it’s complete.

I am unable to import the structured.py of fastai in the kaggle kernel. I think there seems to be an issue with the version of fastai and pytorch in the kernel. It would be very helpful if anyone could point me towards the right direction and help me get rid of this error.

to being able to import structured.py you’ll need to downgrade to fastai 0.7.0 e.g. !pip install fastai==0.7.0